17548636. REVEALING RARE AND ANOMALOUS EVENTS IN SYSTEM AUTOMATION LOGS simplified abstract (INTERNATIONAL BUSINESS MACHINES CORPORATION)
REVEALING RARE AND ANOMALOUS EVENTS IN SYSTEM AUTOMATION LOGS
Organization Name
INTERNATIONAL BUSINESS MACHINES CORPORATION
Inventor(s)
Dave Willoughby of Austin TX (US)
Pavel Kravetskiy of Grafenau (DE)
REVEALING RARE AND ANOMALOUS EVENTS IN SYSTEM AUTOMATION LOGS - A simplified explanation of the abstract
This abstract first appeared for US patent application 17548636 titled 'REVEALING RARE AND ANOMALOUS EVENTS IN SYSTEM AUTOMATION LOGS
Simplified Explanation
The abstract describes a computer-based method, system, and program for classifying a sequence of log entries in a computing system. The method involves preprocessing the log entries and using trained machine-learning systems to predict the likelihood of the next log entry and whether it is unprecedented. These predictions are then combined to determine the classification of the log entries.
- Preprocessing log entries in a computing system
- Predicting the likelihood of the next log entry after a window using a trained machine-learning system
- Predicting whether the next log entry is unprecedented using another trained machine-learning system
- Combining the predictions to classify the sequence of log entries
Potential Applications
- Log analysis and monitoring in computing systems
- Anomaly detection in log entries
- Predictive maintenance in computing systems
Problems Solved
- Efficient classification of log entries in a computing system
- Early detection of unprecedented log entries
- Streamlining log analysis and monitoring processes
Benefits
- Improved accuracy in predicting the next log entry
- Early identification of potential issues or anomalies in a computing system
- Time and cost savings in log analysis and monitoring
Original Abstract Submitted
A computer-implemented method, system, and computer program product for classifying a sequence of log entries of a computing system may be provided. The method may include pre-processing the log entries. The method may also include predicting, as a first output of a first trained machine-learning system, a likelihood of a particular next log entry after the window. The method may also include, predicting, as a second output of a second trained machine-learning system, whether the next log entry is unprecedented. The method may also include combining the first output and the second output for determining a classification of the sequence of log entries.
